Abstract: It is difficult to automatically recognize defects using digital image processing method in X-ray image that tested from lap joint with unequal thickness plates. This attributes to the very different gray levels in background image caused by the unequal thickness of the workpiece, and the continuous change gray levels caused by lap joint. Besides, the position of the lap joint is uncertain placed before tested, i.e. there is always an angel between the directions of lap joint and gravity. And this makes it difficult for defect detection and localization. In this paper, a method of X-ray image correction based on invariant moments is presented to solve the problem of image skew. In addition, a defect detection method combined of image noise suppression, background removal, image segmentation and mathematical morphology is presented. The results show that the proposed method can effectively recognize the gas pores in lap joint with unequal thickness, and it is suitable for automatic detection.